1. Field of the Invention
The present invention relates generally to the fields of gene biochemical pharmacology and drug discovery. More specifically, the present invention relates to method of screening for a compound that inhibits formation of an organism's 5' mRNA cap structure.
2. Description of the Related Art
Processing of eukaryotic mRNA in vivo is coordinated temporally and physically with transcription. The earliest event is the modification of the 5' terminus of the nascent transcript to form the cap structure m7GpppN. The cap is formed by three enzymatic reactions: (i) the 5' triphosphate end of the nascent RNA is hydrolyzed to a diphosphate by RNA 5' triphosphatase; (ii) the diphosphate end is capped with GMP by GTP:RNA guanylyltransferase; and (iii) the GpppN cap is methylated by AdoMet:RNA (guanine-N7) methyltransferase [1].
RNA capping is essential for cell growth. Mutations of the triphosphatase, guanylyltransferase, or methyltransferase components of the yeast capping apparatus that abrogate catalytic activity are lethal in vivo [2-12]. Genetic and biochemical experiments highlight roles for the cap in protecting mRNA from untimely degradation by cellular 5' exonucleases [13] and in recruiting the mRNA to the ribosome during translation initiation [14].
The physical and functional organizations of the capping apparatus differ in significant respects in fungi, metazoans, protozoa, and viruses. Hence, the cap-forming enzymes are potential targets for antifungal, antiviral, and antiprotozoal drugs that would interfere with capping of pathogen mRNAs, but spare the mammalian host capping enzymes. A plausible strategy for drug discovery is to identify compounds that block cell growth contingent on pathogen-encoded capping activities without affecting the growth of otherwise identical cells bearing the capping enzymes of the host organism. For this approach to be feasible, the capping systems of interest must be interchangeable in vivo.
The architecture of the capping apparatus differs between metazoans, fungi, protozoa, and DNA viruses. Metazoan species encode a two-component capping system consisting of a bifunctional triphosphatase-guanylyltransferase polypeptide (named Mce1p in the mouse and Hce1p in humans) and a separate methyltransferase polypeptide (Hcm1p in humans) [6, 9, 15-22]. The budding yeast Saccharomyces cerevisiae encodes a three component system consisting of separate triphosphatase (Cet1p), guanylyltransferase (Ceg1p), and methyltransferase (Abd1p) gene products [7, 10, 11, 23]. In yeast, the triphosphatase (Cet1p) and guanylyltransferase (Ceg1p) polypeptides interact to form a heterodimer [11], whereas in mammals, autonomous triphosphatase and guanylyltransferase domains are linked in cis within a single polypeptide (Mce1p) [18].
Vaccinia virus capping enzyme is a multifunctional protein that catalyzes all three reactions. The triphosphatase, guanylyltransferase, and methyltransferase active sites are arranged in a modular fashion within a single polypeptide--the vaccinia D1 protein [24-30]. Other DNA viruses encode a subset of the capping activities; e.g., baculoviruses encode a bifunctional triphosphatase-guanylyltransferase (LEF-4) and Chlorella virus PBCV-1 encodes a monofunctional guanylyltransferase [31-33]. The guanylyltransferase and methyltransferase domains are conserved between DNA viruses, fungi, and metazoans. In contrast, the triphosphatase components are structurally and mechanistically divergent.
RNA Guanylyltransferase--Transfer of GMP from GTP to the 5' diphosphate terminus of RNA occurs in a two-step reaction involving a covalent enzyme-GMP intermediate [34]. Both steps require a divalent cation cofactor, either magenesium or manganese. EQU E+pppG.rarw..fwdarw.E-pG+PPi (i) EQU E-pG+ppRNA.rarw..fwdarw.GpppRNA+E (ii)
The GMP is covalently linked to the enzyme through a phosphoamide (P--N) bond to the .epsilon.-amino group of a lysine residue within a conserved KxDG element (motif I) found in all known cellular and DNA virus-encoded capping enzymes (FIG. 1). Five other sequence motifs (III, IIIa, IV, V, and VI) are conserved in the same order and with similar spacing in the capping enzymes from fungi, metazoans, DNA viruses, and trypanosomes (FIG. 1) [35].
Hakansson et al. [36] have determined the crystal structure of the Chlorella virus capping enzyme in the GTP-bound state and with GMP bound covalently. The protein consist of a larger N-terminal domain (domain 1, containing motifs I, III, IIIa, and IV) and a smaller C-terminal domain (domain 2, containing motif VI) with a deep cleft between them. Motif V bridges the two domains. Motifs I, III, IlIa, IV, and V form the nucleotide binding pocket. The crystal structure reveals a large conformational change in the GTP-bound enzyme, from an "open" to a "closed" state, that brings motif VI into contact with the .beta. and .gamma. phosphates of GTP and reorients the phosphates for in-line attack by the motif I lysine. When the crystal is soaked in manganese, guanylyltransferase reaction chemistry occurs in crystallo and the covalent enzyme-GMP intermediate is formed. However, only the enzyme in the closed conformation is reactive.
Identification of essential enzymic functional groups has been accomplished by site-directed mutagenesis of Ceg1p, the RNA guanylyltransferase of Saccharomyces cerevisiae. The guanylyltransferase activity of Ceg1p is essential for cell viability. Hence, mutational effects on Ceg1p function in vivo can be evaluated by simple exchange of mutant CEG1 alleles for the wild type gene. The effects of alanine substitutions for individual amino acids in motifs I, III, IIIa, IV, V, and VI [2, 5, 6] have been examined. Sixteen residues were defined as essential (denoted by asterisks in FIG. 1) and structure-activity relationships at these positions were subsequently determined by conservative replacements. Nine of the essential Ceg1p side chains correspond to moieties which, in the Chlorella virus capping enzyme crystal structure, make direct contact with GTP (arrowheads in FIG. 2). These include: the motif I lysine nucleophile which contacts the .alpha.-phosphate of GTP; the motif I arginine and motif III glutamate, which contact the 3' and 2' ribose hydroxyls, respectively; the motif III phenylalanine, which stacks on the guanine base; the two motif V lysines, which contact the .alpha.-phosphate; the motif V aspartate, which interacts with the .beta.-phosphate; the motif VI arginine that interacts with the .beta. phosphate; and the motif VI lysine, which contacts the .gamma. phosphate of GTP [6, 36].
On the basis of sequence conservation outside motifs I, III, lIIa, IV, V, and VI, the capping enzymes of fungi (S. cerevisiae, S. pombe, C. albicans,), metazoans (C. elegans and mammals) and Chlorella virus can be grouped into a discrete subfamily [6, 37]. The sequence alignment in FIG. 2 highlights two motifs that are present in these capping enzymes, but not in the poxvirus enzymes, which can be designated motif P and motif Vc. In the Chlorella virus capping enzyme, motif P forms one wall of the guanosine binding pocket of domain 1 [36]. Motif Vc--(K/R)I(I/V)EC--is situated between motifs V and VI in domain 2. The glutamate residue of motif Vc is essential for the activity of the fungal guanylyltransferase Ceg1p [37].
RNA Triphosphatase--There are at least two mechanistically and structurally distinct classes of RNA 5' triphosphatases: (i) the divalent cation-dependent RNA triphosphatase/NTPase family (exemplified by yeast Cet1p, baculovirus LEF-4, and vaccinia D1), which require three conserved collinear motifs (A, B, and C) for activity [12, 28, 31, 32], and (ii) the divalent cation-independent RNA triphosphatases, e.g., the metazoan cellular enzymes and the baculovirus enzyme BVP [15, 17, 38-40], which require the HCxAGxGR(S/T)G phosphate-binding motif. The existence of additional classes of RNA 5'-triphosphatases is likely, given that the candidate capping enzymes of several RNA viruses and trypanosomatid protozoa lack the defining motifs of the two known RNA triphosphatase families [41, 42]. Hence, the triphosphatase components of the capping apparatus provide attractive targets for the identification of specific antifungal, antiviral, and antiprotozoal drugs that will block capping of pathogen mRNAs, but spare the mammalian host enzyme.
Mammalian RNA Triphosphatase--Metazoan capping enzymes consist of an N-terminal RNA triphosphatase domain and a C-terminal guanylyltransferase domain. In the 496-amino acid mouse enzyme Mce1p, the two catalytic domains are autonomous and nonoverlapping [18]. The metazoan RNA triphosphatase domains contain a (I/V)HCxAGxGR(S/T)G signature motif initially described for the protein tyrosine phosphatase/dual-specificity protein phosphatase enzyme family. These enzymes catalyze phosphoryl transfer from a protein phosphomonoester substrate to the thiolate of the cysteine of the signature motif to form a covalent phosphocysteine intermediate, which is then hydrolyzed to liberate phosphate (FIG. 3). The metazoan capping enzymes hydrolyze the phosphoanhydride bond between the .beta. and .gamma. phosphates of triphosphate-terminated RNA; they are not active on nucleoside triphosphates. The conserved cysteine of the signature motif is essential for RNA triphosphatase function [11, 15, 38, 39]. A characteristic of the cysteine-phosphatases is their lack of a requirement for a divalent cation cofactor.
The N-terminal portion of Mce1p from residues 1-210 is an autonomous RNA triphosphatase domain [18]. Recombinant Mce1(1-210)p has been expressed in bacteria and purified to near-homogeneity. Mce1(1-210)p sediments in a glycerol gradient as a discrete peak of 2.5 S, indicating that the domain is monomeric in solution. The RNA triphosphatase activity of Mce1(1-210)p can be assayed by the release of .sup.32 Pi from .gamma..sup.32 P-labeled poly(A). A kinetic analysis showed that the initial rate of Pi release was proportional to enzyme concentration. Mce1(1-210)p hydrolyzed 1.2 to 2 molecules of Pi per enzyme per second at steady state. RNA triphosphatase activity was optimal in 50 mM Tris buffer at pH 7.0 to 7.5. Activity was optimal in the absence of a divalent cation and was unaffected by EDTA. Inclusion of divalent cations elicited a concentration dependent inhibition of RNA triphosphatase activity. 75% inhibition was observed at 0.5 mM MgCl.sub.2 or MnCl.sub.2. Mce1(1-210) did not catalyze release of .sup.32 Pi from [.gamma..sup.32 P]ATP.
Metal-dependent RNA Triphosphatases--The RNA triphosphatases of S. cerevisiae and DNA viruses are structurally and mechanistically unrelated to the metazoan RNA triphosphatases. The vaccinia virus RNA triphosphatase depends absolutely on a divalent cation cofactor. Vaccinia triphosphatase displays broad specificity in its ability to hydrolyze the .gamma. phosphate of ribonucleoside triphosphates, deoxynucleoside triphosphates, and triphosphate-terminated RNAs [43, 44]. The NTPase and RNA triphosphatase reactions occur at a single active site within an 545-amino acid N-terminal domain of vaccinia capping enzyme that is distinct from the guanylyltransferase active site [27-30]. The vaccinia RNA triphosphatase is optimal with magnesium, is 12% as active in manganese, and is inactive with cobalt [43]. In contrast, the vaccinia NTPase is fully active with cobalt, manganese, or magnesium [43, 44]. Baculovirus LEF-4 hydrolyzes the .gamma. phosphate of RNA and NTPs; the LEF-4 NTPase is activated by manganese or cobalt, but not by magnesium [31].
The yeast RNA 5'-triphosphatase Cet1p also hydrolyzes the .gamma. phosphate of nucleoside triphosphates [12]. The NTPase of Cet1p is activated by manganese and cobalt. This is a property shared with the triphosphatase components of the vaccinia D1 and baculovirus LEF-4 capping enzymes. Recent studies illuminate a common structural basis for metal-dependent catalysis by these enzymes. The metal-dependent RNA triphosphatases share 3 collinear sequence motifs, designated A, B, and C (FIG. 4). These are present in yeast Cet1p, in the Cet1p homolog from yeast Candida albicans, in the triphosphatase-guanylyltransferase domains of the vaccinia virus, Shope fibroma virus, molluscum contagiosum virus, and African swine fever virus capping enzymes, and in baculovirus LEF-4. Mutational analysis identified several residues within these motifs that are essential for the RNA triphosphatase and ATPase activities of vaccinia virus capping enzyme; the essential residues include two glutamates in motif A, an arginine in motif B, and two glutamates in motif C [28]. Alanine substitutions at any of these positions in the vaccinia capping enzyme reduced phosphohydrolase specific activity by 2 to 3 orders of magnitude. These 5 residues may comprise part of the triphosphatase active site. All five residues essential for vaccinia triphosphatase activity are conserved in LEF-4 and Cet1p. Mutations of the equivalent residues of Cet1p result in loss of triphosphatase activity [12].
Physical Association of the Triphosphatase and Guanylyltransferase Components of the Capping Apparatus--Yeast and mammals use different strategies to assemble a bifunctional enzyme with triphosphatase and guanylyltransferase activities. In yeast, separate triphosphatase (Cet1p) and guanylyltransferase (Ceg1p) enzymes interact to form a heterodimer, whereas in mammals, autonomous triphosphatase and guanylyltransferase domains are linked in cis within a single polypeptide (Mce1p).
Cet1p and Ceg1p Form a Heterodimeric Capping Enzyme Complex In Vitro.--The native size of purified recombinant yeast RNA triphosphatase Cet1p was gauged by glycerol gradient sedimentation. Cet1p sedimented as a single monomeric component of 4.3 S. The yeast guanylyltransferase Ceg1p also sediments as a monomer in a glycerol gradient. Yet, when equal amounts of recombinant Cet1p and Ceg1p were mixed in buffer containing 0.1 M NaCl and the mixture was analyzed by glycerol gradient sedimentation, the two proteins, as well as the triphosphatase and guanylyltransferase activities, cosedimented as a single discrete peak of 7.5 S. Thus, Ceg1p and Cet1p interact in vitro to form a heterodimer [11]. Cet1p does not heterodimerize with the structurally homologous RNA guanylyltransferase domain of the mouse capping enzyme. Recombinant mouse guanylyltransferase, Mce1(211-597)p, was purified to homogeneity, mixed with Cet1p or with a buffer control, and then subjected to sedimentation analysis in parallel with the Cet1-Ceg1 mixtures. The 45 kDa Mce1(211-597) protein alone sedimented as a single monomeric peak [18]. Sedimentation of the Mce1(211-597)p plus Cet1p mixture revealed no shift in the distribution of the mouse guanylyltransferase or the yeast triphosphatase to a more rapidly sedimenting form [11]. Hence, yeast RNA triphosphatase forms a heterodimer with yeast guanylyltransferase, but not with the mammalian enzyme. Subsequent studies of the Candida albicans RNA triphosphatase (named CaCet1p) showed that it could interact with S. cerevisiae guanylyltransferase Ceg1p in vivo as gauged by a two-hybrid reporter assay [45].
Cet1p-Ceg1p Heterodimerization is Essential In Vivo--Truncated proteins Cet1(201-549)p and Cet1(246-549)p were expressed in bacteria and purified from soluble bacterial lysates by Ni-agarose and phosphocellulose column chromatography. Purified Cet1(201-549)p and Cet1(246-549)p catalyzed the release of .sup.32 Pi from .gamma..sup.32 P-labeled triphosphate-terminated poly(A) or [.gamma..sup.32 P]ATP with the same specific activity as full-length Cet1p [11]. The CET1(201-259) gene in single copy was functional in vivo in supporting yeast cell growth [11]. The finding that the CET1(246-549) gene on a CEN plasmid could not support cell growth, even though the gene product has full RNA triphosphatase activity in vitro, suggests that the catalytic activity of Cet1p, though essential for cell growth (see below), may not suffice for Cet1p function in vivo. Glycerol gradient analysis showed that Cet1(201-549)p by itself sedimented as a monomer. When Cet1(201-549)p was mixed with Ceg1p and the mixture was analyzed by glycerol gradient sedimentation, the two proteins cosedimented as a 6.8 S heterodimer [11]. The more extensively truncated Cet1(246-549)p did not sediment as a discrete monomer like full-sized Cet1p and Cet1(201-549)p. Rather, most of the Cet1(246-549)p sedimented as a high molecular weight oligomer (.about.13 S) that retained RNA triphosphatase activity. When a mixture of Cet1(246-549)p and Ceg1p was sedimented most of the Cet1(246-549)p remained aggregated; only a minor fraction of the input Ceg1p was shifted to the size expected for a heterodimer. Deletion from residues 201-245 (which results in loss of function in vivo despite retention of triphosphatase activity in vitro) affects the interaction of Cet1p with Ceg1p.
The implication of these data is that the interaction of Cet1p with Ceg1p is essential for yeast cell growth. Pharmacological interference with Ceg1p-Cet1p heterodimerization is a potential mechanism for blocking gene expression in fungi without impacting on mammalian cells.
Cap Methyltransferase--The enzyme RNA (guanine-N7-)methyltransferase (referred to hereafter as cap methyltransferase) catalyzes the transfer of a methyl group from AdoMet to the GpppN terminus of RNA to produce m7GpppN-terminated RNA and AdoHcy [1]. The Saccharomyces cerevisiae cap methyltransferase is the product of the ABD1 gene [7]. ABD1 encodes a 436-amino acid polypeptide. A catalytic domain of Abd1p from residues 110 to 426 suffices for yeast cell growth (8, 9); this segment of Abd1p is homologous to the methyltransferase catalytic domain of the vaccinia virus capping enzyme [7]. A key distinction between the yeast and vaccinia virus cap methyltransferases is their physical linkage, or lack thereof, to the other cap-forming enzymes. The vaccinia virus methyltransferase active site is encoded within the same polypeptide as the triphosphatase and guanylyltransferase, whereas the yeast methyltransferase is a monomeric protein that is not associated with the other capping activities during fractionation of yeast extracts [7]. Mutational analyses of the yeast and vaccinia cap methyltransferases have identified conserved residues that are critical for cap methylation [8, 9, 46]. In the case of Abd1p, mutations that abolished methyltransferase activity in vitro were lethal in vivo.
A putative cap methyltransferase from C. elegans was identified on phylogenetic grounds [9]. An alignment of the sequence of the predicted 402-amino acid C. elegans protein (Genbank accession Z81038) with the yeast cap methyltransferase Abd1p is shown in FIG. 5. Although it remains to be demonstrated that the nematode protein has cap methyltransferase activity, the extensive sequence conservation suggested that other metazoans might also encode homologues of Abd1p. A human cDNA that encodes a bona fide cap methyltransferase has been identified and a physical and biochemical characterization of the recombinant human cap methyltransferase (Hcm1p) produced in bacteria was conducted. A functional C-terminal catalytic domain of Hcm1p was defined by deletion analysis [22].
Interaction of the Cellular Capping Apparatus with the Phosphorylated CTD of RNA Polymerase II. Cap formation in eukaryotic cells in vivo is targeted to the nascent chains synthesized by RNA polymerase II (pot II). A solution to the problem of how pot II transcripts are specifically singled out for capping has been described whereby the cellular capping enzymes are targeted to pre-mRNA by binding to the phosphorylated carboxyl-terminal domain (CTD) of the largest subunit of pol II [16-18, 47, 48]. The CTD, which is unique to pol II, consists of a tandem array of a heptapeptide repeat with the consensus sequence Tyr-Ser-Pro-Thr-Ser-Pro-Ser. The mammalian pol II large subunit has 52 tandem repeats whereas the S. cerevisiae subunit has 27 copies. The pol II largest subunit exists in two forms, a nonphosphorylated IIA form and a phosphorylated IIO form, which are interconvertible and functionally distinct. In vivo, the pol IIO enzyme contains as many as 50 phosphorylated amino acids (primarily phosphoserine) within the CTD. During transcription initiation, pol IIA is recruited to the DNA template by the general transcription factors. The pol IIA CTD undergoes extensive phosphorylation and conversion to IIO during the transition from preinitiation complex to stable elongation complex. Several CTD kinase activities have been implicated in CTD hyperphosphorylation, each of which contains a cyclin and cyclin-dependent kinase subunit pair. The cdk7 and cyclin H subunits of the general transcription factor TFIIH catalyze phosphorylation of Ser-5 of the CTD heptapeptide. Other CTD kinases include the cdk8/cyclin C pair found in the pol II holoenzyme, CTDK-I, a heterotrimeric kinase with cdk-like and cyclin-like subunits, and P-TEFb, a regulator of polymerase elongation with a cdc2-like subunit.
The recombinant S. cerevisiae and Sc. pombe guanylyltransferases Ceg1p and Pce1p bind specifically to the phosphorylated form of the CTD [16]. Moreover, recombinant yeast cap methyltransferase Abd1p also binds specifically to CTD-PO4 [16]. Phosphorylation at Ser-5 of the heptad repeat was sufficient to confer guanylyltransferase and methyltransferase binding capacity to the CTD [16]. This analysis has been extended to mammalian capping enzyme where the key finding is that the guanylyltransferase domain Mce1(211-597)p by itself binds to CTD-PO4, whereas the triphosphatase domain Mce1(1-210)p does not [18]. These findings suggest that the mammalian RNA triphosphatase is targeted to the nascent pre-mRNA by virtue of its connection in cis to the guanylyltransferase. The phosphorylation-dependent interaction between guanylyltransferase and the CTD is conserved from yeast to mammals. It is not clear if the structural elements on the yeast and mammalian enzymes that interact with CTD-PO4 are conserved or divergent. Nonetheless, pharmacological interference with the binding of guanylyltransferase or cap methyltransferase to the CTD is a potential mechanism for blocking gene expression in fungi or mammalian cells. Such interference can occur either by direct blocking of capping enzyme/CTD-PO4 binding or indirectly by affecting the phosphorylation state of the CTD.
The prior art is deficient in the lack of methods of screening for a compound that inhibits formation of an organism's 5' mRNA cap structure. The present invention fulfills this longstanding need in the art.